We have tried to compare the results of V22.04.00.40 & V22.05.00.131 prior to shift to the latest update - same STAAD file (no modifications).
As shown in the snapshot, significant post-processing output differences were found.
The largest differences noted were on the Wind Load Cases.
The STAAD Generated Wind Load values, however, were found to be the same.
So it must be a change in the calculation/routine that caused the significant difference.
Can anyone please clarify the wind load calculation procedure for Update 5?
Could you please attach the related STAAD.Pro model?
I have uploaded the file through the Communities Secure File Upload.
I have uploaded the excel file I used for comparison. Basically I just got the post-processing results from the analysis of both versions, no other modification.
After further investigation it appears that this difference exists due to a defect introduced in STAAD.Pro version 22.05.00.131 where some modifications to the TRUSS specification were implemented. More details can be found here: https://communities.bentley.com/products/ram-staad/f/ram-staad-forum/205770/new-staad-version-22-05-00-131-giving-different-results/624523#624523
As far as I can see, this issue will be addressed in the upcoming STAAD.Pro release.
Meanwhile, changing TRUSS spec to member releases should give the same results between both versions.
To further clarify, since attached link was focused on temperature load, the bug is actually due to the TRUSS specification, correct?
Can you also inform when the next version release would most likely be?
Thanks!
Yes, the issue is related to TRUSS specification and some loadings on it. The current plan is to release the next version in March.
I tried the member release workaround you suggested, I also removed the Temperature load to eliminate its effect.
However, I still get significant differences for the Wind Load Cases ONLY. (snapshot is for the Beam End Forces output).
I have sent another excel file with tabs for the trials I have made.
a) No Truss specification, only member release.
b) No Temperature Load but with Truss Specification
c) No Truss Specification, No Temperature Load.
I think besides the TRUSS Specification bug, the issue here is really with the wind load routine. Please advise.
It seems that wind load generated between both models is slightly different. I will try to figure out why this happens. However, is a difference of about 0.3% really so significant?
It's the effect of this difference in the output that is significant especially under Direct Analysis. Members that passed in Update4 are now failing in Update5. We are still taking a look at what else could have changed but your help would be great to hasten the investigation. This is actually the first time we encountered significant differences in the output of STAAD versions.
It sounds strange that a difference of 0.3% in bending moment could change the design results significantly. How exactly are the design results different between 2 versions (when the model is without TRUSS spec)?
After further investigation I noted that there is a slight difference in the wind loads generated on the members between different versions. It looks like some wind loads were not correctly generated on version 22.04.00.40. I will check this with the development team and get back to you.
I just checked the primary loads for the cases mentioned since the trial item was big, but most likely the results are not far from what I originally reported. And for all the cases done, the significant differences was consistently ONLY in the Wind Load Cases. I do hope we can clarify this matter soon. Thank you!